The intercristal (Tuffier’s) line is a clinical landmark defined as a horizontal line connecting the superior aspect of the posterior iliac crests. It is routinely used to locate the level of the body of L4 prior to performing a subarachnoid or epidural block. Identifying a particular intervertebral level for neuraxial techniques by clinical palpation is inaccurate.1 , 2 Reports of spinal cord injuries and neurologic sequelae from neuraxial techniques based on Tuffier’s line have been noted.3 Magnetic resonance imaging (MRI) and radiography studies highlight possible explanations for the unanticipated injuries incurred at presumed “safe” levels of needle insertion, including (1) differences in the vertebral level of the palpated intercristal line vs the imaged (or anatomic) intercristal line4; and (2) significant interpatient variation of the vertebral level at which spinal cord termination (conus medullaris) occurs.5

As ultrasound technology becomes more accessible to the clinician, it holds great potential to facilitate neuraxial anesthesia for the anesthesiologist and patient. In the obstetric population, ultrasound reliably predicts pre-puncture lumbar epidural space depth,6 enhances patient satisfaction with the epidural insertion process, and decreases needle insertion attempts.7 Moreover, ultrasound improves accuracy in selecting a particular vertebral interspace vs palpation-based techniques.8 To our knowledge, ultrasound has not been reported to assess the vertebral level of the palpated intercristal line in the non-obstetric population. This cross-sectional study uses ultrasound to determine the vertebral level at which the palpated intercristal line occurs in volunteers clinically positioned for neuraxial techniques. Based on previous studies assessing palpated intercristal line as measured by radiography,4 we hypothesized that the level of the palpated intercristal line would correspond to the L3-4 vertebral interspace in the majority of subjects, as measured by ultrasound.

Methods

The Ottawa Hospital Research Ethics Board approved both the study and the advertisements used for recruitment. Informed, written consent was obtained from all participants. Study subjects consisted of pre-admission unit patients and allied healthcare staff at the Ottawa Hospital. Recruitment was facilitated through advertisements located throughout the Ottawa Civic Hospital. Exclusion criteria included age <18 yr, pregnancy, and inability to maintain clinical positioning for a neuraxial technique.

The age, height, weight, and waist circumference of the participants were directly measured by a single observer (C.P.), and body mass index (BMI) was then calculated. With volunteers in clinical position for neuraxial blockade (seated on a stool with feet on a footrest and neck, hips, and back flexed), the superior aspect of the iliac crests was palpated from behind. A horizontal skin mark was created with an erasable pen at the level of the palpated iliac crest. A SonoSite® MicroMAXX® (Bothell, WA, USA) ultrasound using a C60e 2–5 MHz 60-mm broadband curved array probe was applied in a paramedian orientation to the spine. Following sonographic identification of the sacrum, the ultrasound probe was advanced cephalad, counting the vertebral interspaces until the middle of the probe intersected with the horizontal skin marking of the palpated intercristal line. The interspace at that level was determined to be the ultrasound-measured palpated intercristal line. If the palpated intercristal line intersected the cephalad half of the spinous process, the intervertebral level cephalad to the spinous process was recorded. Similarly, if the palpated intercristal line intersected the caudad half of the spinous process, the intervertebral level caudad to the spinous process was recorded. If the palpated intercristal line intersected the middle of the spinous process, the intervertebral level caudad to the spinous process was recorded.

The primary objective of the study was to use ultrasound to characterize the location of the palpated intercristal line. The secondary outcome of the study was to investigate physical features (sex, age, height, mass, BMI, and waist circumference) as potential predictors of intercristal line location. On the basis of a previous study,4 we expected palpation of the intercristal line to identify the L3-4 interspace in 75% of participants. We considered a 95% confidence interval with expected half-width ≤50% to be a sufficiently precise estimate for this proportion, yielding a required sample size of approximately 100. As such, this was our target sample size.

Subjects were divided into three groups based on the lumbar interspace identified on ultrasound (L2-3, L3-4, and L4-5). To ensure a family-wise Type 1 error rate of 5% for each characteristic of interest we used Tukey’s Honestly Statistically Different (HSD) test to make pair-wise comparisons of the physical characteristics (age, height, mass, BMI, and waist circumference) between the groups. Pearson’s Chi square test with a Bonferroni adjustment of the resulting P values (3 × P) was used to make pair-wise comparisons of the sex ratio between groups. No additional adjustment was employed to account for the multiple variables compared between the groups.

Results

One hundred and fourteen consenting subjects (64 females and 50 males) were entered in the study. The palpated intercristal line was found at the L3-4 interspace in 83 of 114 (73%; 95% confidence interval 64% to 80%) subjects. Demographic characteristics are displayed in Table 1. Males were taller, heavier, and had greater waist circumferences than females. The age of the volunteers ranged from 24 to 88 yr and did not differ between sexes. Individuals with a palpated intercristal line at the L2-3 interspace were more frequently male (Table 2) and taller (Table 3). Table 3 highlights the relationship identified between physical characteristics and intercristal line height.

Table 1 Physical characteristics of subjects
Full size table
Table 2 Distribution of palpated intercristal line levels by sex
Full size table
Table 3 Physical characteristics of subjects with different palpated intercristal line levels
Full size table

Discussion

This cross-sectional analysis study used ultrasound to assess the vertebral level of the palpated intercristal line in volunteers positioned for a neuraxial technique. Our findings are consistent with recent clinical studies evaluating the palpated level of the intercristal line by radiography,4 i.e., the majority of subjects have a palpated intercristal line at L3-4.

Commonly used anesthesia texts cite the intercristal line as “crossing the body of L4 or the L4-5 interspace”9 , 10 and do not differentiate between the palpated and imaged intercristal line when describing performance of a subarachnoid or epidural block. This difference is important to consider as the palpated, not the imaged, intercristal line forms the basis for initiating neuraxial attempts in clinical practice. Given that up to 10% of the population has a spinal cord proper extending caudal to L2,5 the potential for hazard from erroneously placed needles based on inaccurate surface anatomical relationships exists and has manifested in patient morbidity.3 Despite the risk of nerve injury with subarachnoid approaches, a recent study has demonstrated that permanent neurologic sequelae are extremely rare (0.9–1.2 cases per 100,000) and supports the overall safety record of central neuraxial anesthetic techniques.11

We have potentially identified a clinically important subset of patients in whom the palpated intercristal line occurred at L2-3, i.e., males and taller individuals. Our results parallel work by Kim et al. who used MRI to demonstrate a more cephalad position of the imaged intercristal line in males compared with females.12 A recent publication by Snider et al. also supports our findings that the intercristal line is more cephalad in males than in females.13 Spino-pelvic morphological dissimilarity between males and females may account for the observed difference in vertebral level of the palpated intercristal line. Legaye et al. identified that the pelvic incidence angle—“the angle between the perpendicular to the sacral plate at its midpoint and the line connecting this point to the middle axis of the femoral heads”—is significantly greater in males than in females.14 Subsequently, Horduna and Legaye reported a highly significant, positive correlation between the magnitude of the pelvic incidence angle and a more cephalad intercristal line location by radiography.15 Taken together, these studies may provide an anatomic explanation for the greater proportion in our study of males-to-females with a more cephalad ultrasound-confirmed palpated intercristal line.

Fifteen subjects (13.2%) had their palpated intercristal line at the L2-3 interspace. This cephalad location of the intercristal line raises the possibility of needle placement in close proximity to the termination of the spinal cord should a clinician assume that the intercristal line corresponds to the L3-4 interspace. Subjects with a palpated intercristal line sited at L2-3 were more frequently male (Table 2) and taller (Table 3) than those sited at L3-4 or below. As male subjects were taller than females, we considered the use of logistic regression to determine if both height and sex were independent predictors of cephalad placement of the palpated intercristal line. Peduzzi et al. state that logistic regression models require a minimum of ten events for each variable entered into the model to prevent overestimation of effects.16 We had considered sex, height, and weight as candidate variables a priori requiring a minimum of 30 instances of a palpated intercristal line identified at L2-3 for regression modelling. As there were only 15 cases of a palpated intercristal line identified at L2-3 by ultrasound, logistic regression was not performed. To our knowledge, little research has been conducted investigating the role of patient height and sex of the patient in determining the level of the palpated intercristal line. Further studies are warranted to assess the possible risk to which taller and/or male patients may be exposed during neuraxial techniques.

It is important to note the limitations of our study. First, a single investigator assessed both the palpated intercristal line and ultrasound-measured vertebral levels. The lack of blinding introduces the potential of bias when recording results. Second, we did not document the incidence in our study population of lumbo-sacral transitional vertebrae (LSTV), commonly referred to as “sacralization of the L5 vertebra” or “lumbarization of the S1 vertebra”. Given that the incidence of LSTV has been reported as high as 5–15%,17 a similar proportion of participants in our study who may possess this anatomical aberration may affect the accuracy of ultrasound confirmation of the palpated intercristal line. Although Kim et al. showed with MRI that sacralization patients have more cephalad imaged intercristal lines and coni medullari compared with lumbarization or non-LSTV patients, there was no difference in the “margin of safety”—distance between the intercristal line and the conus medullaris—among subjects with sacralization, lumbarization, or normal anatomy.12

In conclusion, this cross-sectional study used paramedian neuraxial ultrasonography to assess the intervertebral level at which the palpated intercristal line occurred in non-obstetric subjects clinically positioned for a neuraxial technique. We found that the majority of subjects had an ultrasound-confirmed palpated intercristal line at L3-4. However, further research is required to determine the importance of physical factors, particularly height and sex, when considering location of needle placement for a neuraxial technique.